The Valuation of Cash Flow Forecasts: An Empirical Analysis

Abstract

This article compares the market value of highly leveraged transactions (HLTs) to the discounted value of their corresponding cash flow forecasts. For our sample of 51 HLTs completed between 1983 and 1989, the valuations of discounted cash flow forecasts are within 10 percent, on average, of the market values of the completed transactions. Our valuations perform at least as well as valuation methods using comparable companies and transactions. We also invert our analysis by estimating the risk premia implied by transaction values and forecast cash flows, and relating those risk premia to firm and industry betas, firm size, and firm book-to-market ratios. THIS ARTICLE COMPARES THE market value of management buyouts and leveragedrecapitalizations to the discounted value of their corresponding cash flow forecasts. Most economists readily accept the concept of estimating market values by calculating the discounted value of the relevant cash flows. However, little empirical evidence exists that shows that discounted cash flows provide a reliable estimate of market value. This study provides evidence of a strong relation between the market value of the highly leveraged transactions (HLTs) in our sample and the discounted value of their corresponding cash flow forecasts.Our tests compare the transaction values in HLTs to estimates of the present value of the relevant cash flows. We use a sample of management buyouts and leveraged recapitalizations because these transactions typically release the cash flow information and transaction value required for the analysis. We use the cash flow forecasts to estimate the cash flows that will accrue to all capital providers, including different classes of debt and equity. The Journal of FinanceWe estimate a terminal value when the cash flow information ends. We value the capital cash flows and the terminal value using a discount rate based on the Capital Asset Pricing Model (CAPM). The resulting median estimates of discounted cash flows are within 10 percent of the HLT transaction values. Furthermore, the prediction errors of the discounted cash flow estimates (relative to transaction values) are qualitatively similar to those found in previous work for option pricing models (relative to call option prices).We compare the performance of our discounted cash flow estimates to that of estimates obtained from alternative valuation approaches that rely on companies in similar industries and companies involved in similar transactions. Such alternative valuation approaches-known as comparable or multiple approaches-are commonly used in practice. The discounted cash flow (DCF) methods, individually, perform at least as well as the comparable methods. However, we also find the comparable approaches to be useful, especially when combined with a discounted cash flow valuation.The discounted cash flow methods we use generally parallel the basic techniques taught in most business schools. Our results suggest that those techniques are both useful and reliable. We stress that our valuations rely...

Strategic Implications of Valuation: Evidence from Valuing Growth Options

“…Consistent with finance theory and practice, we model the residual value as a perpetuity. We assume a 4% growth rate for the cash flows to equity occurring after 1999, which is the rate Kaplan and Ruback (1995) suggest as an appropriate growth rate. Negative residual values occur when the 1999 cash flow to equity is negative.…”

“…If the intrinsic value is less (greater) than market value, the ratio will be less (greater) than 1 and the log ratio will be negative (positive). Kaplan and Ruback (1995) suggest the use of the log ratio because it -…is symmetric with respect to overestimates and underestimates‖ (1995: 1070). We perform two-tailed t-tests to determine whether the valuation errors are significantly greater/less than zero.…”

“…Specifically, we use the log of the ratio of the intrinsic (DCF) value to market value, similar to Kaplan and Ruback (1995). This log ratio represents a measure of the valuation error of the DCF model.…”

“…First is the assumed growth rate for the cash flows to equity used in the full 10-year, multi-period DCF model. We use 4%, the rate Kaplan and Ruback (1995) use, which allows for a modest growth rate for the firm after inflation. Lower or higher rates could have been used, but 4% was in line with standard research practice and seemed reasonable.…”

A Bayesian framework for combining valuation estimates

“…DCF is implemented by forecasting expected cash inflows from operations, netting them against forecasted payments to creditors, and then discounting. The discount factor, which depends on interest rates, market risk, and leverage, is estimated from CAPM and the weighted average cost of capital (Kaplan and Ruback 1995). In a certain world with perfect markets, DCF is cut and dry.…”